To explore the application of neural networks for estimating geomagnetic field disturbances, this study pays particular attention to K-index classification. The primary goal is to develop a robust and efficient method for classifying different levels of geomagnetic activity using neural networks. Our work encompasses data preprocessing, model architecture optimization, and a thorough evaluation of classification performance. A new neural-network approach is proposed to address the specific complexities of geomagnetic data, and its merits are compared with those of conventional techniques. Notably, Long Short-Term Memory (LSTM) models significantly outperformed traditional methods, achieving up to 98% classification accuracy. These findings demonstrate that neural networks can be effectively applied in geomagnetic studies, supporting AI-based forecasting and enabling further integration into space weather research

In each magnetic observatory, the magnetic field is registered in local Cartesian coordinate systems associated with the geographic coordinates of the locations of these observatories. To observe extraterrestrial magnetic field sources, such as the interplanetary magnetic field or magnetic clouds, a method of joint processing of data from magnetic observatories of the international Intermagnet network was implemented. In this method, the constant component is removed from the observation results of individual observatories, their measurement data is converted into the ecliptic coordinate system, and the results obtained from all observatories are averaged after the coordinate transformation.The first data on joint processing of measurement results from the international network of Intermagnet magnetic observatories in the period before the onset of magnetic storms of various types, during these storms and after their end are presented. There is a significant improvement in the signal-to-noise ratio after combining the measurement results from all observatories, which makes it possible to isolate weaker external magnetic fields. A change in the shape of magnetic field variations is shown, which can provide new knowledge about the mechanism of development of magnetic storms.

The formation of international networks of magnetic observatories has provided new opportunities for monitoring geomagnetic disturbances and solving problems related to space weather. On the other hand, collective processing of data from numerous observatories revealed problems with the quality of data provided by some of them. Sometimes, the reason for the decrease in data quality is associated with faulty measuring instruments. However, every year, industrial interference also increases its influence on measurement quality. INTERMAGNET suggests carrying out a mutual comparison between the values of neighboring magnetic observatories as one of the methods for solving the problem of maintaining data quality. This paper proposes a method to improve the quality of magnetograms, which makes it possible to eliminate the results of interference of local origin through the use of four spatially separated magnetic observatories located in Kazakhstan. This method is suitable for automating the magnetograms’ quality control process.

The integration of magnetic observatories of many countries into the Intermagnet network has provided researchers with new opportunities to investigate many geophysical processes. The joint processing of measurement evidence from magnetic observatories located in various geolocations makes it possible to better understand the dynamics of the interaction of solar plasma fluxes with the Earth's magnetosphere. For the correct interpretation of the dynamics of physical processes, the accuracy of the time reference of measurement results is of great importance. The measurement results of local observatories were transformed to a unified ecliptic coordinate system oriented towards the Sun in order to extract small variations in the magnetic field occurring before the magnetic storm commencement. Before transforming the coordinates, a constant component equal to the average daily value of the magnetic field components was removed from the results of minute-by-minute measurements. Owing to the averaging evidence from 84 magnetic observatories of the Intermagnet network, variations in the magnetic field of cosmic origin, preceding the magnetic storm commencement were detected. The correlation analysis of these variations recorded by different observatories revealed the presence of time shifts between the data series of different observatories. Only 62% of the observatories had a relative data shift within the range of ± 1 minute. Before the joint data processing in order to efficient use of data recorded with a time shift a method for correcting time shifts has proposed.